DESCRIPTION (provided by One can readily envision many applications for the unnatural base pairs applicant): The diversity of all living organisms is encoded within their DNA, where it is stably maintained through DNA replication, and then retrieved through transcription into RNA and translation into proteins. Thus, the diversity of life is limited by the four natural nucleotids that comprise DNA and RNA and the twenty natural amino acids that they encode. Drawn by the potential conceptual and practical ramifications, chemists and biologists have long been fascinated by the idea of expanding the genetic alphabet. This requires an unnatural base pair that is replicated and transcribed, both efficiently and with high fidelity and without any significant sequence bias. Expansion of the genetic alphabet would make possible the site-specific labeling of DNA and RNA - just as these biopolymers are receiving increased attention for applications ranging from novel materials to therapeutics. The ability to synthesize or evolve DNA/RNA that is site-specifically modified with functional groups of interest, and thus has activities outside the scope of their natural counterparts, promises to greatly increase their potential applications. Expansion of the genetic alphabet would also lay the foundation for the first semi-synthetic organism, able to store increased information in its DNA and retrieve it in th form of proteins with unnatural amino acids. In the previous funding period we developed the first unnatural base pair, d5SICS-dNaM, that is replicated and transcribed with efficiency and fidelity approaching that of a natural base pair and without any significant sequence bias. Molecular recognition within this pair is based not on complementary hydrogen- bonding, as with the natural base pairs, but on complementary hydrophobic and packing forces, more similar to the forces underlying protein structure and folding. We also developed a polymerase selection system to identify mutants that better recognize the unnatural base pair. With these tools in hand, we now propose to use a variety of synthetic and biological methods to: 1) Complete the characterization of the structural determinants of d5SICS-dNaM stability, replication, & transcription; 2) Complete the optimization of (d)5SICS and (d)MMO2/(d)NaM for replication, transcription, and site-specific labeling of DNA and RNA; 3) Demonstrate the utility of the expanded genetic alphabet for multiple in vitro applications; 4) Optimize T7 phage DNA polymerases for the in vivo replication of DNA containing d5SICS-dNaM; and 5) Initiate efforts to expand the genetic alphabet in vivo by establishing a system in E. coli for the uptake of (d)NaMTP and (d)5SICSTP. The completion of these aims should elucidate the mechanisms underlying DNA replication, and make available general tools for the site-specific labeling of DNA and RNA, as well as tools for the evolution of polymerases with desired activities. Completion of the proposed work will also position us at the doorstep of creating the first semi-synthetic organism, able to store, retrieve, and evolve, with more information than natural organisms.